Compare models

Pick up to 4models. Specs render side-by-side. Share the URL — it's stateless.

Comparing agents instead? Switch to agent compare →

Selected

Gemini 3.5×

Qwen3.6-27B×

DeepSeek-R1×

DeepSeek-V4-Pro×Clear all

	Gemini 3.5 Google	Qwen3.6-27B Alibaba	DeepSeek-R1 DeepSeek	DeepSeek-V4-Pro DeepSeek
Vendor	Google	Alibaba	DeepSeek	DeepSeek
Family	Gemini	Qwen	DeepSeek	DeepSeek
Release date	2026-05-20	2026-04-22	2025-01-20	2026-04-22
Context window	2,000,000 tokens	262,144 tokens	128,000 tokens	1,000,000 tokens
Parameters	—	27B (dense)	671B	1.6T (49B active)
Modality	text, vision, audio, video	text, vision, video	text	text
License	proprietary	Apache-2.0	MIT	MIT
Source	proprietary	open weights	open weights	open weights
Description	Google DeepMind's next-gen Gemini — positioned as "frontier intelligence with action". Built for complex agentic workflows. Announced at Google I/O 2026.	Alibaba's first dense open-weight in the 3.6 family. Strong agentic-coding scores (77.2 SWE-bench Verified, matching Claude 4.5 Opus on Terminal-Bench 2.0). Supports 201 languages and multimodal text/image/video input.	Reasoning-focused open-weight model.	DeepSeek's flagship open-weight MoE. 1.6T parameters with 49B activated, 1M-token context, and a hybrid attention scheme (CSA + HCA) that delivers long-context inference at ~27% of V3.2's FLOPs.
Links	vendor →	weights →	weights →	weights →
Benchmarks
MMLU-Pro	88.9%	78.9%	—	84.2%
GPQA-D	91.2%	—	—	82.4%
HumanEval	—	91.5%	—	95.1%
Aider	—	70.2%	—	80.1%
AIME-25	96.1%	—	79.8%	88.6%
LiveCB	—	—	—	72.4%
MMMU	86.0%	70.3%	—	—